Biomaterial section for OBI manuscript

[Melanie Courtot]

Hi all,

The biomaterial paragraph for the manuscript is available at
http://docs.google.com/Edit?docid=dzprnmw_80fp52wkdd
It has already been extensively edited (thanks!), so feel free to
review and update it as well.

I will send it to Bjoern on behalf of the branch on Wednesday morning.

A summary is below, note that the google doc contains comments from
reviewers that I didn't copy here.

Thanks,
Melanie


The OBI Biomaterial branch was named after the MGED Biomaterial term.
In the context of OBI, this branch is meant to encompass material
entities of biological origin and chemical molecules (for example,
water) and deal with different granularity levels (atoms, organism,
populations). These lead us to create a class material_entity by
declaring the union of the different types of BFO Objects
(fiat_object_part, object and object_aggregate), and use this class as
our root instead, deprecating the biomaterial class. MaterialEntity
has now been moved into the upper ontology BFO. Addition of
appropriate restrictions on our classes allows reasoning and inference
of a more structured hierarchy.

The Biomaterial branch contains classes that have been heavily
described by other ontologies. In order to maintain orthogonality to
other resources and avoid duplication of efforts, the mechanism of
Minimal Information to Represent an External Ontology Term (MIREOT)
has been implemented. The MIREOT approach was initially tested on the
Cell use case, allowing replacement of the OBI Cell with an import of
the Cell term from the Cell Ontology.

While all elements are important in the Biomaterial branch, some were
deemed more pivotal. Extra work has been taken to define them, add
appropriate logical constraints, and coordinate integration with other
OBI branches. The core Biomaterial elements are described below:

1. organism: Considering that lots of time and resources had already
been dedicated to building taxonomies, we chose to reuse the NCBI
taxonomy and therefore declare organism as being the union of Viruses,
Bacteria, Archae and Eukaryota. In addition, OBI doesn't take position
on whether or not sperm, fetus and eggs are to be considered
organisms.

2. organization: the definition and placement of organization has been
subject to a lot of controversy. An organization has been declared as
a MaterialEntity in OBI. Legal status of organizations is left open
for discussions and out of scope for OBI. The organization class will
be used and instantiated (e.g. Affymetrix instance) as needed.

3. molecular entities : this class and its subclasses are being
imported (via the MIREOT mechanism) from ChEBI. When an addition is
needed, we submit a request onto the ChEBI tracker for addition. Some
issues were encountered when larger additions were needed. For example
the Flow Cytometry community has a list of 600 fluorochromes to add,
and the Immunology community would like to insert 1000 peptides. These
large additions will be held into OBI and stored in separate files.
End-users are allowed to decide if any of these extra-modules is
needed.

4. chemical entities in solution: this class is generated to encompass
things like buffers. To efficiently define these terms, the
concentration of one entity in another is needed. For example, glucose
concentration is a molecular concentration of glucose in a solution.
Measurement of glucose concentration is_about some glucose
concentration and scalar measurement datum and has measurement unit
label and some concentration unit label. One of the instance of the
measurement of glucose concentration class is "my measurement of
glucose concentration", which would have properties
has_measurement_unit_label mg/ml and is quality measurement of my
blood sample (an instance of a blood sample), and has measurement
value 1.25.

5. anatomical entity (aka macroscopic part of organism): this class is
generated to represent all anatomical parts of the organism. It is
expected to be a placeholder for importing terms from FMA and CARO.
This class, however, excludes granular parts of organism, like atoms,
cells. generally everything that can be removed from the organism
without affecting it. Since the distinction micro/macroscopic is
sometimes hard, this class anatomical entity is renamed to clearly
mean an identifiable structure in the organism.

6. processed material: this class is at the interface of the
Biomaterial and Instrument branches. It covers everything that is
manufactured (e.g., devices), extracted body part (e.g., extracted
heart, and blood sample), collection of material, and synthetic
biological elements. It is defined as being the output of a processing
material process.

In addition to the development choices mentioned above, some other
cases are still problematic in the Biomaterial branch. For example, it
has proven very difficult to find a satisfactory way of defining an
entity of organismal origin (see http://docs.google.com/Doc?docid=dzprnmw_79d5fh9qdn&hl=en).
Some work also needs to be done in coordination with EnvO to properly
define environmental matter.